Introduction
Column 1 {data-width = 650}
Corpus
Sitting in a cafe or restaurant, music has a big influence on the atmosphere. Working in a cafe myself, I use a lot of spotify playlists. In this cafe we have different playlists for different moments of the day. For my corpus in this research I chose our lunchtime and dinnertime playlists. I want to find out what aspects of this music makes them more enjoyable on different parts of the day. For this I will compare these two playlists with each other. Beforehand, I would expect differences mostly in the tempo and energy of the tracks.
- beschrijven van verschillen tussen playists. En verwachte verschillen.
To investigate the differences I make use of a variation of visualisations and spotify features. First I will visualize track-level features for songs in both playlists. Secondly I will analyse the chroma features of both playlists by using chromagrams.
Column 2 {data-width = 350}
Volgende
Visualisation
Column 1
First graph
For the first plot, I put danceability and energy on the axes. I thought these were two good features to see the difference between the parts of the day.
Chromagrams
Column lunch
lunch playlist
Column evening
evening playlist
Loudness
Timbre
Tempo
Hw week 10 - what is this
For these visualizations I chose two songs that are very representative for the playlists. I think you see very good that in the evening playlist, there is a lot more variation in the keys used. This could be explained by the lunch playlist being more calm and equal, so also the difference in keys being more equal.
lunch
lunch playlist
evening playlist
Evening playlist
Dendogram
Column graph
Graph
Column explanation
Explanation
This dendogram is made with the lunch playlist. In the dendogram you can see that there are several clusters to be found. However, due to the large amount of data it is not very readable yet. I am thinking of reducing the number of songs used. Maybe by creating a playlist with songs from both lunch as evening playlists. In this way, I could maybe use the dendogram to see if based on features, the songs from the same playlists are clustered together.